Quote from: Priyanka on August 14, 2013, 12:34:48 PM
I am not debating either of these points. I greatly admire anyone who spends time in studying and understanding the game and believe there is a lot to learn from them. The thing am getting at is someone need not give things on a plate, but atleast let us know the strategy and thought process behind what they do (just like Superman explained his money management) it will benefit eager people like who is trying to learn and understand the game. If it cannot be explained, then it cannot be.
My whole point was around botting stuff. What I was getting to was if it is situation awareness and making decisions based on what you see and what you have experienced, it can be taught to a machine and such artificial intelligence can be coded. It is not that its an easy job, but being in the robotics industry, I know exactly how its done. But if its purely based on human instincts, then it cannot be.

Let me take the crude example into account. The key word is "Factor into account". The moment that comes in, it becomes more of a decision making rather than a voodoo capability. Hope you are seeing where am getting to.

I did my training at Ensemble studios before graduating and starting work as a R&D consultant with Fanuc and Universal robotics. During the training, I experienced how we build AI capabilities into the RTS game Age of empires. It is simply an  astonishing subject to look at when it comes to situational awareness and taking 1000s of factors into consideration before taking the next decision which lasts only for milliseconds. And it also has inbuilt learning capabilities based on what it experiences and based on 100s of decisions happened in the past.

It is more like building a Deep blue. But remember, Deep blue was indeed beaten by Kasparov.

Thank you for a very interesting reply Priyanka.

I was always kind of undecided if more complex strategies could be botted. You have certainly convinced me that it can be done. It also makes you wonder what kind of research the casino industry has done involving bots because of the resources they have and the time to do it in.

Number Six

There is a book called Fortune's Formula by William Poundstone which goes into great depth on the Kelly criterion.

It is a good book because it charts the beginnings of the formula and the people who used it.

Copies are easily available on Amazon for anyone interested.

cheers.

Instinct kind of downplays it a bit IMO. Players like Superman have spent a lot of time studying the game and doing the groundwork. I liken it more as a kind of situational awareness. A players results over time would likely show losses if they were just using pure instinct.

A player could walk up to a table with an instinct that number 23 was about to appear. Unbeknown to the player could be the fact that the 23 area is part of a cold section on the wheel due to a temporary bias. Another player at the table may have taken this factor into account and have an instinct that number 26 is going to come up on the other side of the wheel. Ask yourself who is more likely to profit in the longer term. I suppose this is a more extreme example using a physical approach, however it does illustrate a point.

You don't win longterm using voodoo or guesswork.

The scatter on those reels makes it unplayable IMO. [smiley]aes/tongue.png[/smiley]



P.S. For anyone interested. There is a guy over on the Wizard of Vegas Forum who has started posting recently called 'mickeycrimm'
His stories are fascinating and he claims to be an AP player at slots who has lived in the Vegas area for years. I am pretty certain these are not the kind of slots you will find in the U.K. or Oz. These are like Video Poker etc.. But it is great reading and you never know what you might pick up from the guy.

cheers

[attachimg=1]

Quote from: Bayes on August 05, 2013, 07:14:18 PM

Think about the so called "law of the third" which says that after 37 spins there will be 12-13 numbers which haven't hit. That means that every 37 spins there is a group of numbers equivalent to a dozen or column (at least) which are still sleeping, but systems based around betting on those sleepers to wake up haven't been very successful.

That's true Bayes. I have been there and bought the T-shirt! (several of them)

Roy Ward Dickson might have something where he waits for one of these sleepers to appear and then starts attacking them in the hope that they will compensate from their absence with a short burst of wins. (I will start a separate topic with some of his ideas)

cheers

Creating bets from dozens and columns is one of my favourite ways of playing roulette.

Why do I say that?

Because dozens and columns can sleep. This creates variance. You need variance if you are a flat bettor and hope to profit in a session.

What I am going to show you is something that I can't really say that I have seen anyone else mention on any of the forums.

The first aim is to find out what is the sleeping dozen and column.

I will type out a few examples....

23 2B (So this is pretty self explanatory. 23 belongs to the second dozen and column B)
36 3C (1A) (36 belongs to the third dozen and column C. So the sleeping dozen is the 1 and the sleeping column is the A. Hence the 1A at the end in the brackets)

Another example....

09 1C
06 1C
11 1B
34 3A (2C)

Another....

33 3C
27 3C
10 1A (2B)

So you get the idea. The sleeping dozen and column could also be referred to as the furthest back dozen and column.

Now what I have found quite often is that one of either the sleeping dozen or column will appear on the next spin.

So taking the most recent example above....

33 3C
27 3C
10 1A (2B)
19 2A (3B)  (You can see how the sleeping second dozen appeared here)

Let's grab another number....

33 3C
27 3C
10 1A (2B)
19 2A (3B)
30 3C (1B)  (The sleeping third dozen appeared here)

Now here is what I have noticed....

You can see above how the sleeping dozen keeps changing. However the sleeping column keeps sleeping.

So how would you take advantage of this?

Look at the number 10 in the above example. Let's say you want the second dozen to appear and the column B to remain sleeping.

So you would be looking for the second dozen to appear. But it can only appear in either column A or Column C if the B column is to remain sleeping.

So that's going to be either 2A or 2C.

2A is represented by the 13/16 split and the 19/22 split.

2C is represented by the 15/18 split and the 21/24 split.

You can see that the number appearing after the 10 was 19 and this belongs to one of the 4 splits above.

So the new sleeping dozen and column after the 19 came was 3B.

So repeating the process....

I am looking for the column B to remain sleeping. However I am going to bet for the third dozen to appear.

So what I am looking for is either 3A or 3C.

3A is represented by the 25/28 split and the 31/34 split.

3C is represented by the 27/30 split and the 33/36 split.

The next number out was 30 belonging to the 3C.

The new sleeping dozen and column is now 1B.

Like I said, you will find this happens a lot. You do get a change in the sleeping dozen and column. But we also know that dozens and columns can sleep for a long time. So this means that sometimes you are going to get just either the sleeping dozen or column changing continuously leaving the other to sleep. This allows you to play just 4 splits.

One thing I have noticed with dozens and columns is how the variance tends to come and go in waves.

So you will get something like....

28 3A
11 1B (2C)
07 1A (2C)
28 3A (2C)
31 3A (2C)
04 1A (2C)   (So you can see here how the 2C continues to sleep as the missing dozen and column. Obviously you would not bet what I am describing above when you see this happen. Just be patient and wait for one of either the sleeping dozens or columns to start appearing)

Here is an example of that....

09 1C
28 3A (2B)
14 2B (1C)
15 2C (1A)
22 2A (1B)
32 3B (1C)   (You can see here how the sleeping dozen and column keep changing every spin)

It does not matter if it does not change every so often.

You might get something like....

change
change
not change
change
change
change
not change

This is ok just getting the odd intermittent no change. You can work through this.

Where you don't want to betting is....

not change
not change
change
not change
not change
not change
change

Like I said, I have noticed these trends seem to be easier to read and are more pronounced when working with dozens and columns. It is not difficult to get the hang of it.

cheers

Hello Turner

Go to rouletteforum.net home page.

Click the link where it says 'old public forum'

Then the first section is 'Roulette forum / Free roulette systems.... click that.

Then it is a sticky! The 6th one down.

I am pretty certain Poit did a few of these and one was for more than 3 million spins and covered a lot of different stuff.

I will have a look myself later and let you know if I find anything else.

cheers

I found your stats interesting Turner

Poit on the old rouletteforum.net did a 3 million spin test. I tried to copy + paste it here, but I can't paste anything in this forum for some reason.

Here is a pretty simple idea that may be worth testing a bit to see how it goes.

I will give a quick example using the sixlines and the quads.

Sixlines...

1-6
7-12
13-18
19-24
25-30
31-36

Quads...

1-9
10-18
19-27
28-36

I got the following numbers from Spielbank Wiesbaden.

24
13
31
19
27
29
24
4
18
29
30
14
22
29
17

I will convert the above numbers to the Quads first.

3
2
4
3
3
4
3
1
2
4
4
2
3
4
2

Now it's pretty simple. All you do is look for the most recent two of the same and bet for that until there is a change and you have a new most recent two.

So here is how it would go for the Quads.

3
2
4
3 **The 3 here is the most recent two**
3  win (+3)
4  loss (+2)
3  win (+5)
1  loss (+4)
2  loss (+3)
4  loss (+2)  **The 4 is now the most recent two**
4  win (+5)
2  loss (+4)
3  loss (+3)
4  win (+6)
2  loss (+5)  **The 2 is now the most recent two**


Now here are the numbers converted to the Sixlines.

4
3
6
4
5
5
4
1
3
5
5
3
4
5
3

Here is how it would go for the Sixlines.

4
3
6
4 **The 4 is the most recent two**
5  loss (-1)
5  loss (-2)  **The 5 is the most recent two**
4  loss (-3)
1  loss (-4)
3  loss (-5)
5  win (level)
5  win (+5)
3  loss (+4) 
4  loss (+3)
5  win (+8)
3  loss (+7)  **The 3 is the most recent two**

That's it basically.

I was thinking about progressions for this.

For the Quads...

Maybe something like 111 222 333

For the Sixlines...

1111 2222 3333

or something along those lines.

cheers

Betting against sleepers is one way to try and keep the variance as low as possible to allow you to have a profitable session.  The house edge won't kill you in a single session, where as the variance most certainly can.  So betting against anything sleeping for a bit is naturally going to give you some kind of winning run if it continues to sleep.

The 64 dollar question is what is the optimal bet location for betting against sleepers?

My personal opinion is things like betting against a sleeping dozen or column (meaning you back the other two) is not a long term winner. A few losses can quickly dig a hole which may be hard to get out from. The best way I have found is to limit the bet location to as few numbers as possible.

But here is the crux of the problem really. How can you limit the bet and keep variance at bay. Solve that equation and you are making some progress. The good news for us is that loads of things can sleep for a long time. So it pays to be creative and flexible in looking for these sleeping locations.

Quote from: D1 on August 03, 2013, 10:01:47 PM
The top 5 numbers add up to 69 ?

The bottom 5 numbers add up to 92

There is a duck in there as well! Feathers everywhere. [smiley]aes/confused.png[/smiley]

Here is the first group of numbers today from Spielbank Wiesbaden on Table 3.

[attachimg=1]

That looks like a nice win right there on the 1,3,5,7,9 group.

EDIT: You are sitting at +726 after that last number 7 appeared. The next 4 losses would swallow up 290 chips still leaving the profit at +436.

Quote from: soggett on August 02, 2013, 06:31:16 AM


btw - so you play only one color at a time?not both?

soggett, there is nothing stopping you playing both sides if you want to. However, it does get a bit expensive on the losing cycles. This is very streaky IMO. So it makes sense to look for a dominant side after a quite period. It might be that both sides wake up from a slumber and start performing well. That would be even better.

I don't find it that difficult to track this just watching the marquee. So it doesn't really need to be someone's primary method.

But I think it is brilliant used as an attacking ploy if you spot the opportunities. No harm in risking 40-50 chips every so often in an attempt to hit the jackpot.

cheers

Hello

Here is a more detailed example of the way I was playing last night using this concept.

31 black appeared and was then followed by 33.

So the group 7 consisting of numbers 29,31,33 and 35 was playable.

I am going to go up 1 chip on each number on every spin. (win or lose)

Number 7 appeared and that's a loss of 4 chips.

So now I am going up to 2 chips on each number.

Number 29 appeared. That's a win and I am ahead by 60 units.

Now I am going up to 3 chips on each number.

Number 34 appeared and that's a loss. Down now to +48 units.

Now I am going up to 4 chips on each number.

Number 16 appeared and that's a loss. Down now to +32 units.

Now I am going up to 5 chips on each number.

Number 29 appeared. That's a win and I am ahead by 192 units.

(Here is the thing! Your next 3 bets would be 6, 7 and 8 chips on each number. That's 84 chips. That would still leave you 108 units ahead should you lose the next three bets.)

So now I am going up to 6 chips on each number.

Number 36 appeared. That's a loss and now the profit is 168 units.

Now I go up to 7 chips on each number.

Number 35 appeared. That's a win and I am ahead by 392 units.

Now I go up to 8 chips on each number.

Number 33 appeared. That's a win and I am ahead by 650 units.

Now I go up to 9 chips on each number.

Number 7 appeared. That's a loss and now the profit is 614 units.

Now I go up to 10 chips on each number.

Number 21 appeared. That's a loss and now the profit is 574 units.

(So obviously you have to decide at this stage how long you are prepared to chase another winner. You have plenty of chips in the bank so to speak. However the bets are slowly escalating every spin and another 4/5 losses from here is going to wipe out around 250 units. I explained in the above post how I think giving a number 2/3 chances to keep appearing seems like the best way to go.)

Now I go up to 11 chips on each number.

Number 29 appeared. That's a win and the profit is now 926 units.

It's fair to say this was a pretty hot streak and obviously this won't happen all the time. The good news is they don't need to.
On saying that, look at the example from one of the posts above where there was 19 wins vs 3 losses. That session or one like it could seriously win someone a lot of money.

The way I approach this is to mark both sides (red and black) and wait for a cold period. Then I am looking for it to wake up again. It sometimes wakes up with a bang and that's what I like about the up 1 unit on a win or loss staking plan.

cheers